W. R. VAUGHAN,R. L. CRAVEN, R. Q.LITTLE,JR.,
-A
e----
\
\
\
\
230
250 270 290 310 Wave length, mp. Fig. 1.-Ultraviolet spectra of: A, 2-phenylcyclopentanone; B, 2-phenylcyclohexanone (adapted from W. C. Wildman and R. B. Wildman, J . Org. Chem., 17, 581 (1952)) ; C, 3-phenyl-2-butanone (adapted from W. D. Kumler, et a1.9 ; solvent 95% ethanol. violet and infrared spectraIsg the ketone was freshly purified aiu the semicarbazone.a The ultraviolet spectrum (0.02 (39) In this connection, t h e assistance afforded b y correspondence with Dr. Ralph Nusbaum and his staff, Spectroscopy Section, Atomic Energy Project, U.C.L.A., Los Angeles, Calif., is gratefully acknowledged. (40) On prolonged standing, both 2-phenylcyclopentanone (W.
[CONTRIBUTION FROM
THE
AND
A. C. SCHOENTHALER VOl. 77
u t . yo solution) is reproduced in Fig. 1; it is characterized by a shelf near 280 mp, three shoulders in the 250-270 mp region and a broad absorption band below 250 mp. The infrared spectrum features a strong carbonyl band a t 5.75 p . Resolution of 2-Phenylcyc1opentanone.-The resolving agent was prepared's by conversion of meth2l (+)-tartrate ~ into methyl (+)-tartramide, m.p. 132-135 , [ a ] +57.8' ~ followed by con(water) (lit.15m.p. 136-140°, [ a ]62.9"), version into the (+)-tartramazide, m.p. 147-148', [ L Y ] * ~ D +122' (water) (lit.'5 m.p. 146-148", ID 124'). A solution of 43 g. (0.26 mole) of (+)-tartramazide and of 36 g. (0.225 mole) of 2-phenylcyclopentanone in 100 ml. of methanol was refluxed for three hours. The oil which separated on standing could not be induced to crystallize A solution of the oil in 3: 1 methanol-chloroform, on standing in the refrigerator, deposited a crop of crystals. This material was recrystallized from the same solvent mixture ten times, until melting point and rotation had reached a constant value. There was thus obtained 0.75 g. of,(+)-2phenylcyclopentanone ( +)-tartramazone, m.p. 151 , [a]"D +183.1' (c 4.98, methanol). Anal. Calcd. for ClaHISO4X3: ?T, 13.76. Found: N, 13.64. The above tartramazone was added to a well-stirred refluxing mixture of ligroin and dilute hydrochloric acid. The ligroin layer yielded 0.2 g. of ( +)-2-phenylcyclopentanone, Rotations4' are recorded in Table I V . m.p. 25-27'. Anal. Calcd. for CllH120: C, 82.5; H , 7.55. Found: C, 82.4; H , 7.42.
Acknowledgment.-A grant-in-aid from Research Corporation is gratefully acknowledged. Baker and P. G. Jones, J. Chem. Soc., 787 (1951))and Z-phenylcyclohexanone (W. E. Bachmann, G I. Fujimoto and I,. B Wick. ' T i m JOURNAL, 7 2 , 1995 (1950))are known t o deteriorate. (41) No claim for optical purity can he made, hut it seems reasonable t o suppose t h a t t h e method employed in the hydrolysis of the tartramazone should be conducive t o a s negligible a degree of racemization as was observed in t h e hydrolysis of 2-phenyl-R-pentanoiie semicarbazone, under t h e same conditions.
N E W YORK53. hT.Y.
DEPARTMENT O F CHEMISTRY O F
THE
UNIVERSITY O F MICHIGAS]
Addition of Hydrogen Halides to a#-Unsaturated Acids] B Y WYMAN
R. V A U G H A N , ROBERT L. CRAVEN,^ RANDEL Q. LITTLE,J K I 3AND A. CHARLES SCIIOENTHALER RECEIVED MAY19, 1954
Evidence for the trans addition of hydrogen halides to a,@-unsaturatedacids is reviewed and new evidence is presented t o show that in general the addition occurs by a trans mechanism, which cannot involve formation of a simple enolic intermediate or a classical carbonium ion.
It has been assumed generally that hydrogen halides add to olefinic and acetylenic linkages in a trans sense. It is our purpose to establish the generality of this assumption as factual for olefinic acids in which the double bond is conjugated with the carboxyl group. Fittig has reported that tiglic and angelic acids both afford the same h ~ d r o b r o m i d e but , ~ the hydroiodides of these acids differ, and upon dehalogenative decarboxylation, respectively, yield transand &butene5 as would be expected from trans hy-
drohalogenation followed by trans dehalogenative decarboxylati~n.~-~ In the same sense a-ethylcrotonic acid adds hydrogen iodidelo and hydrogen bromidel' t o give hydrohalides both of which afford trans-2-pentene on treatment with base. Likewise cis-2-methyl-3-ethyl- and trans-2-methyl-3-ethylacrylic acids afford, respectively, trans- and cis2-pentene by way of their hydroiodides.'? The formation of pure meso-2,3-dibromosuccinic acid by hydrobromination of bromomaleic acid would be convincing evidence for trans hydrobromination
(1) AbstractFd in the main from portions of the Ph.D. Dissertations of Robert L. Craven (1954) and Randel Q. Little, Jr. (1953), University of Michigan. (2) American Brake Shoe Company Fellow 1951-1952. .4lhert B. Prescott Fellow in Organic Chemistry 1953. (3) American Brake Shoe C o m m n y Fellow 1952-1953. f i ) R. Fittig and A. Pugenstrecker, Airn., 196, 108 (1870). L5) W. G . Young, R. T, Dillon and H. J. Lucas, THISJOURNAI.,61, 252s (1929).
(6) W. R. Vaughan and K. M. Milton, ibid., T4, 5623 (1952). (7) S.J. Cristol and W. P. Xorris, ibid,,T 6 , 632,2645 (1953). (8) E. Grovenstein and D . E. Lee, ibid., 7 6 , 2639 (1953). (9) R. L. Craven, P h . D . Dissertation, University of Michigall, 1954. (IO) M. L. Sherrill and E . S. Matlack, THIS J O U R N A L , 6 9 , 2134 (1937). (11) W. M. Lauer and F. H. Stodola, ibid.. 66, 1216 (1934) ( 1 2 ) 11 J . Lucas aud A N. I'ruter, ; b i d , , 59, 1682 (1937).
March 20, 1953
ADDITIONOF HYDROGEN HALIDESTO UNSATURATED ACIDS
1595
if the same paper did not report that bromofumaric acid also affords the same product, almost to the exH o , C ~ o * H clusion of the expected racemic form.13 Clearly experimental evidence concerning the stereochemistry of the hydrohalogenation of a,@-unsaturated acids is suggestive of trans hydrohalogenation but is inadequate to permit any generalization; the 111, X = C1 V, X = H single unequivocal case is the hydrobromination of dibenzo [2,2,2]bicycloiictatriene-2,3-dicarboxyliclowed by cis addition; (B) acid-catalyzed enoliacid (I) in which only the product of trans addition zation, with essentially irreversible loss of water to is formed.a give the anhydride 11; (C) reverse Diels-Alder, followed by isomerization of chlorofumaric acid CO,H I dehydration to the anhydride and readdition to anthracene; (D) bimolecular (SN2) attack by chloride ion resulting in halide exchange-epimerization. This need not be considered for I11 because of the well-established hindrance to intermolecular nucleophilic displacement for this type of I system.6 For VI1 (below) this mechanism is less The recent report that hydrogen bromide adds rigorously excluded even though potassium brotrans to simple mono ole fin^^^ is encouraging, but mide is not very effective in epimerizing VZI. Preowing to the simpler system involved, i t is,not pos- sumably this is attributable to the relatively low sible to transfer conclusions to the olefinic acid sys- concentration of bromide from the potassium salt as compared with 30% hydrogen bromide. tem without reservations. A number of reactions were run in an attempt to I n connection with other studies in this Laboratory i t seemed desirable to repeat the work of establish the nature of the isomerization, and the Vaughan and Miltona using hydrogen chloride. results are given in Table 1. By way of ascertaining When the reaction was attempted in glacial acetic if elimination of hydrogen chloride was essential acid, saturated a t 0" with hydrogen chloride, no to the isomerization process, the deschloro analogs reaction occurred after 12 hours, but a t 135' for I V and V of I1 and 111, respectively, were prepared three days 2-chlorodibenzo [2,2,2]bicyclooctadiene- from anthracene, maleic anhydride and fumaric cis-2,3-dicarboxylic anhydride(I1) was obtained in acid. Since reactions 4 and 5 (Table I) represent inter92.2% yield; apparently cis addition had occurred. rupted runs under identical conditions and for identical times, the similarity in yields suggests that the isomerization and dehydration of I11 to 11 (reaction 3) and the apparent cis hydrochlorination of I to I1 proceed through a common ratecontrolling stage. The yield (60%) of I1 from anthracene and chloromaleic anhydride is signifi11, X = C1 IV, X = H cantly higher than in 4 and 5 , and appreciable anThe investigations of Cristol and his collaborators thracene was recovered owinq to the comparative on the dehydrochlorination of a closely related sys- slowness of the reaction. Thus since the Dielstem16 suggest that, since cis dehydrochlorination is Alder reaction is first order in both addends,16-17 encountered a t elevated temperatures, the stereo- one would expect an appreciable build-up of anchemistry of hydrochlorination should be cis. thracene in reactions 4 and 5 if process C were inThe same work predicts that trans elimination TABLE I should dominate a t low temperatures, and consequently hydrochlorination should be trans, as for HYDROCHLORINATION OF I AND EPIMERIZATION Sol- Temp., Time, ProdYield, hydrobromination,Oand in agreement we find that Run Reactant vent" OC. days uct % after eight days a t 25" in 35% hydrogen chloride 1 I a 135 3 I1 92.2 in glacial acetic acid a 67.7% yield of 2-chlorodiI b 25 8 I11 67.7 benzo [2,2,2]bicyclooctadiene-trans-2,3-dicarboxylic 2 I1 98.0 3 I11 a 135 3 acid (111) is obtained. a 135 6hr. 11 36.1 4 I11 It is possible that addition a t both temperatures 5 I a 135 6 h r . I1 36.4 is trans, and that I1 is formed by isomerization of 6 111 c 135 3 I 52b 111. The trans-acid I11 therefore was subjected 7 Fumaric acid and a 135 3 IV 33 4 to conditions under which I afforded 11, and quananthracene V 32.8 titative conversion occurred. Four processes may 8 V a 133 3 IV 40.5 be considered for this isomerization : (A) EliminaV 52 2 tion of H X (only trans elimination is possible) fol(13) M. S. Kharasch. J. V. Mansfield and F.R . Mayo, unpublished results; cf. 1 . V. Mansfield, Ph.D. Dissertation, University of Chicago, 1948. (14) G. S. Hammond and T. D. Nevitt, THIS JOURNAL, 76, 4121 (1954). (15) S. J. Cristol and N. L. Hause. ibid., 74, 2193 (1952).
a Solvents: a , hydrogen chloride-acetic acid, saturated at 0 ' ; b, hydrogen chloride-acetic acid, 37%; c, glacial acetic acid. Based on chloride ion.
(16) A. Wasserman, J . Chcm. Soc., 828 (1935). (17) W. E. Bnchmann and L. B. Scott, THISJOURNAL, 70, 1458 (1948).
1596
IT.R. VAUGHAN, R. L. CRAVEN,R. Q. LITTLE,JR.,
volved in the isomerization. Since no anthracene was detected, process C may be discounted. Support for either -4 or E, or a combination of the two, is adducible from reaction 8 which shows that an Unfavorable equilibrium exists with respect to 111; and this also may explain the failure to isolate 111 from any but the most highly concentrated hydrogen chloride solutions. The failure to isolate anthracene from the isomerization of V (reaction S), which actually afforded more I V than did reaction 7, again implies noninvolvement of process C ; and since reaction X cannot involve process A, it becomes evident that process B is of primary importance for isomerization. However, the fact that quantitative isomerization of TI1 occurs (reaction 3) under conditions where less than 507, isomerization of V is observed (reaction 8) suggests that process A may be involved where it is possible. Alternatively one must invoke the generalization that a "substituted anhydride system is less readily opened (i.e., is more stable) than the corresponding unsubstituted system in IV, thus admitting of a considerably less labile equilibrium in reaction 3. This generalization is justified by repeated observations in this Laboratory that 2- and 2,3-substituted derivatives of IT or IV show increasing reluctance toward anhydride opening with increasing size of the substituents.'* If this hypothesis is accepted, one may state that the isomerization proceeds principally by process €3. In order to determine this more definitely, a study was undertaken with cyclohexenecarboxylic acid (VI ) whose hydrohalides might be interconvertible, but where the constitution of the equilibrium mixture would be governed by the thermodynamic stabilities of the cis-VI1 and trans-VI11 forms without the complicating factor of subsequent anhydride formation. Both the 2-chloro and 2bromo epimeric pairs were prepared, and, owing to the greater ease of preparation, the bromoacids were selected for study. The trans isomers were prepared from the appropriate trms-2-haloacrylic acids and butadiene, followed by hydrogenation ; and each acid melted lower than its corresponding epimer, although the difference in melting points for the cic- and fvans-bromoacids was only a few degrees. In order to obtain better evidence several attempts to equilibrate the bromoacids by heating each in the presence of bromide ion or hydrogen bromide were made. The results were not absolutely conclusive; however, comparatively little change was observed with the trans isomer while the cis isomer was altered extensively, as revealed by the melting points of the recovered acids. Both the cis-chloro- and cic-bromoacid were prepared by hydrohalogenation of cyclohexenecarboxylic acid ( V I ) . Since the principal product in each reaction was not identical with the trans isomers as prepared above, no doubt exists as to the configurations and, consequently, no doubt exists as to the stereochemistry of hydrohalogenation. Analysis of a residue from one reaction, in which VI was shown to be absent, indicated that a certain (18) For a reported example see W R Vaughan and R Q Little, J r , THISJ O U R N A L , 1 6 , 4130, footnote 9 (1954).
.4ND
-4. c. SCHOENTHALER
Vol. 77
amount of the trans isomer also was obtained during the hydrohalogenation reaction, but no entirely satisfactory method has been developed for estimating the exact relative quantities of cis- and trans-2-bromocyclohexanecarboxylic acids in such a mixture. An attempt to use infrared spectroscopy was abortive, since no clearly characteristic bands could be differentiated in synthetic mixtures. Xccordingly, approximations based on a rough melting point-composition chart were used. In any case it appears that a t moderate temperatures with a large excess of hydrogen bromide (in glacial acetic acid) some 82% of pure cis isomer is produced (trans hydrobromination) , thus leaving less than 38% of the trans isomer in the recrystallization residues. Several crops of different melting point ranges were collected, and by comparison with the melting points of synthetic mixtures, it appears that there is not less than 40yo and not more than Z070 of the trans isomer in the residues. The composition of the residues is mhat might be expected from a crude melting point-composition diagram which indicates a eutectic of this approximate composition. Thus the extent of apparent cis addition is 35-3 8%. In a reaction run a t 10@110" and using a smaller excess of hydrogen bromide there was obtained a 44% yield of pure cis isomer, and a 32% yield of mixed isomers was obtained as a second crop. If the latter is assumed to be about 50% trans, as suggested by its melting point range, this means that relatively more trans isomer is produced a t the higher temperature. It remains, then, to account for the formation of the frans isomer during hydrohalogenation (for the same situation appears to exist with hydrochlorination where only of cis isomer is isolable). If one accepts the premise that the trans isomer is the more stable thermodynamically, as seems to be indicated by the extent of isomerization of the cic and trans forms under epimerizing conditions, the hypothesis that the composition of the mixture is thermodynamically controlled may be discounted. There must be considered, then, process A (which seems improbable since dehydrobromination of VI1 does not occur to any extent in refluxing acetic acid) and a modification of process B, omitting the subsequent anhydride formation. Since relatively less pure VI1 is recoverable after attempted epimerization under conditions approximating those for its formation from V I than is isolable from a preparative run, one may infer that such VI11 as is formed in the preparative run results from epimerization and that the relatively lower quantity is a consequence of never having as high a concentration of VI1 during its actual preparation. Thus process R seems to be responsible for the product of apparent c i r hydrobromination. Inferentially this conclusion may be applied to the bicyclic system as well, with the proviso that the more extensive isomerization of IT1 as compared with V be attributed to the relative stabilities of the two anhydride systems Finally it may be stated that hydrohalogenation of an olefinic bond conjugated with a carboxyl group proceeds by an essentially stereospecific mechanism in which the hydrogen and halogen be-
March 20, 1955
ADDITIONOF HYDROGEN HALIDESTO
UNSATURATED ACIDS
1597
come attached to opposite sides of the carbon-carbon a-bond. Control of the process is mechanistic rather than purely thermodynamic, but epimerization, a t a slower rate, of the hydrohalide initially formed mAy occur to an extent governed by thermodynamic principles. The intermediate in Jolved during such an epimerization is probably enolic, since no appreciable VI11 is formed from VI1 and bromide ion; and since any rational representatio:~of such a structure must admit of nearly equal probability of protonation from either side, i t cannot be an intermediate in the hydrohalogenation process. By the same argument, initial protonation of the carbon-carbon a-bond to give a classical carbonium ion is excluded. The evident strong dependence of hydrohalogenation on the concentration of hydrogen halide suggests a greater than first-order dependence on this reactant (cf. reference 14), and thus i t is not profitable to speculate with regard to the exact nature of the mechanism without kinetic evidence, and there is no reason to suppose that such evidence as has been presented for non-polar solvents1D can be transferred to reactions in glacial acetic acid. However, by ruling out an enolic intermediate during hydrohalogenation, it becomes conceivable that the mechanism may be the same for all olefinic bonds, whether or not14 they are conjugated with a carboxyl group.
2-Chlorodibenzo[2,2,2]bicycloijctadiene-t~ans-2,3-dicarboxylic Acid (III).-A mixture of 2.01 g. (0.00688 mole) of I, 30 ml. of glacial acetic acid and 19 g. of hydrogen chloridt was sealed in a tube a t the temperature of liquid air and allowed to stand a t room temperature for eight days. After allowing the excess hydrogen chloride to evaporate, 0.08 g. of an insoluble inorganic material was separated. The filtrate was treated with Norit and evaporated to a partially crystallized oil, which completely solidified on drying overnight in a vacuum desiccator; yield 2.38 g., m.p. 192-198' dec. Crystallization from nitromethane gave 1.53 g. (67.6%) of 111, m.p. 215-216" dec. Two recrystallizations from nitromethane raised the melting point to 217.5218.5' dec. Anal. Calcd. for Cl8Hl~C1O4:C, 65.76; H, 3.98; C1, 10.79. Found? C, 65.79; H , 4.03; C1, 10.54. The methyl ester formed by the reaction of I11 with diazomethane was recrystallized from methanol two times; m.p. 141-142'. Anal. Calcd. for C~oHl~C104: C, 67.32; H, 4.80; CI, 9.94. Foundz3: C, 67.39; H , 4.90; C1, 9.75. Comoarative Rate of Formation of I1 from I and 111. (A) By' Incomplete Hydrochlorination of 1.-A solution .of 0.444 g. (0.00152 mole) of I and 25 ml. of glacial acetic acld, saturated with hydrogen chloride a t O', was heated in a sealed tube a t 135' for six hours. After allowing the excess hydrogen chloride to evaporate, the solution was treated with Norit and filtered. From the filtrate 0.J47 g. (31.7%, corrected 36.4%) of I1 (m.p. 148.5-153.0 , mixed m.p. 150.5-154.5') was isolated from the neutral fraction. (B) By Incomplete Isomerization of 111.-A solution of 0.500 g. (0.00152 mole) of I11 and 25 ml. of glacial acetic acid, saturated with hydrogen chloride a t O", was heated in a sealed tube a t 135' for six hours. After allowing the excess hydrogen chloride to evaporate, the solution was treated with Norit and filtered. From the filtrate 0.146 g. (30.9%, corrected 36.1%) of I1 (m.p. 150.0-153.5', mixed Experimental20aqb m.p. 151-155') was isolated. Dibenzo[2,2,2]bicycloijctatriene-2,3-dicarboxylic Acid Thermal Decomposition of 111.-A solution of 0.501 g. (I).-This compound was prepared according to the direc- (0.00152 mole) of 111 in 25 ml. of glacial acetic acid was tions of Diels and Alder.2' heated in a sealed tube a t 135" for three days. Nitrogen 2-Chlorodibenzo[2,2,2]bicycloijctadiene-cis-2,3-dicar- was passed through the reaction mixture and into a trap conboxylic Anhydride (11). A.-This substance was prepared taining aqueous silver nitrate and 0.113 g. (0.000788 mole, according to the directions of Synerholm.22 Under condi- 51.8%) of silver chloride separated. tions identical to those for runs 4 and 5 (Table I) a 60% After evaporating the reaction mixture to dryness, it was yield was obtained, whereas longer heating afforded a extracted with two 30-ml. portions of boiling water. From 90.9% yield. the cooled water extracts was obtained 0.171 g. of a white B. By Addition of Hydrogen Chloride to 1.-A solution solid which softened to a semi-solid a t 210' and melted a t of 2.00 g. (0.00684 mole) of I and 100 ml. ofoglacial acetic 230-236" dec. By refluxing 0.146 g. of this material with acid, saturated with hydrogen chloride a t 0 , was heated 1.5 ml. of absolute methyl alcohol and 0.1 ml. of concenin a sealed tube a t 135' for three days. After allowing the trated sulfuric acid for seven hours, 0.117 g. of the dimethyl excess hydrogen chloride to evaporate, the solution was ester of I was obtained, m.p. 152.5-156.0'; mixed with an treated with Norit and filtered. The partially crystallized authentic sample (157.8-159.5') it melted a t 155.5-158.0' oil obtained by evaporating the filtrate was triturated with (previously reportedzl 160-161 "). a few ml. of toluene and dried in a vacuum desiccator; 2.09 trans-2-Chlorocyclohexene-4-carboxylic Acid.-Four g., m.p. 147.5-150.5' to a cloudy liquid. Fractional crys- tubes were each charged with 5.8 g. (0.11 mole) of tank tallization from toluene yielded 1.35 g. (63.5%) of 11, butadiene, 20-30 mg. of hydroquinone and 2.13 g. of transm.p. 154.0-155.3". No depression occurred in melting 6-chloroacrylic acid24 in 30 ml. of benzye. The tubes were point when it was mixed with an authentic sample (155.0sealed and heated for 30 hours a t 150 . After opening the 156.5') of 11. I n addition, there was obtained 20 mg. of tubes, the slightly colored contents were removed and the impure I, 0.31 g. (14.5%) of less pure I1 and 0.13 g. of un- solvent distilled off under reduced pressure. The rather identified high melting solid. gummy, slightly yellow solid residue weighed 14.61 g. This C. By Isomerization of 111.-A solution of 1.00 g. residue was dissolved only partly on treatment with 200 (0.00304 mole) of I11 and 50 ml. of glacial acetic acid, satuml. of boiling petroleum ether (40-60'). The undissolved rated with hydrogen chloride a t O " , was heated in a sealed solid was filtered off, washed with petroleum ether and tube a t 135' for three days. After allowing the excess hydried; 7.003 g., m.p. 108-110", remained. Concentration drogen chloride to evaporate, the solution was treated with of the filtrate by boiling yielded, on cooling to room temNorit and filtered. The partially crystallized oil obtained perature, 1.980 g., m.p. 108-110"; continued concentraby evaporating the filtrate was triturated with a few ml. of tion gave 0.455 g. of gummy solid, m.p. 90-109°. A retoluene and dried in a vacuum desiccator; yield 0.926 g. crystallizoation of the latter material yielded 0.145 g., m.p. (98.0%), m.p. 149.5-153.0'. Recrystallization from tolu108-110 Attempts t o recover more product by recrysene afiorded pure 11. I1 can be recovered in 86% yield tallization from petroleum ether and from water or by subfrom a mixture of I, I1 and I11 approximating the compo- limation of the very oily residue gave only small quantities sition obtained in C. of very impure solid. The latter was not included in the total yield which was 9.128 g., 71%. A second petroleum (19) F. R . Mayo and J. 3. Katz, THISJOURNAL, 69, 1339 (1947); ether recrystallization narrowed and raised the melting F.R Mayo and M. G. Savoy, i b i d . , 69, 1348 (19411). point range to 108.6-109.8' cor. Further recrystallizations (20) (a) Melting points are uncorrected unless otherwise noted. If did not improve upon this value.
.
corrected, the value is exact to f 0 . 5 " . (b) Glacial acetic acid used was Baker and Adamson reagent; where 30% hydrogen bromideacetic acid was used it was Eastman Kodak Co.,White Label, 1161. (21) 0. Diels and K. Alder, A n n . , 486, 191 (1931). (22) M. E. Synerholm, THIS JOURNAL, 61, 1229 (1945).
(23) Microanalysis by Clark Microanalytical Laboratory, Urbana, Ill. (24) E . Gryszkiewicz-Trochimowski, W. Schmidt and D. Gryszkiewia-Trochimowski, Bztll. soc. chim., 693 (1948).
1598
w.R.
VAUGHAN,
R.L.
CR4VEN,
R.
9. LITTLE,JR.,
AND
A.
c. SCHOENTHALER Vol. 77
Anal. Calcd. for C7HpC102:C, 52.35; H, 5.65. Found? A sample of this compound mixed with an equal amount C, 52.54; H , 5.69. of the trans isomer melted a t 60-90". cis-2-Bromocyclohexannecarboxylic Acid (VIIIb).-This trans-2-Bromocyclohexene-4-carboxylic Acid.-Utilizing the same procedure as in the case of the chloro acid, 3.9 g. compound was prepared using a modification of Aschan's procedure2@from cyclohexenecarboxylic acid27,28and 30% (0.072 mole) of butadiene and 2.023 g. (0.0134 mole) of trans-8-bromoacrylic acid26 were sealed in a combustion hydrogen bromide in acetic acid. VIIIh was found t o tube with 50 ml. of benzene and a trace of hydroquinone. melt a t 111.2-112.2", reporteda 112'. Examination of the Low Melting Fractions.-A sample of The tube was heated for 23 hours a t 150". After opening the colored oily residue (m.p. 50-80") obtained from the the tube, transferring the almost colorless homogeneous solution to a flask and removing the solvent in an air stream mother liquors from recrystallization of VIIIh was titrated a t room temperature, the white, slightly gummy crystalline with bromine for unsaturation. Assuming any olefin presresidue was dried in a vacuum desiccator. The crude prod- ent to be unreacted starting material, the amount of the latter was not greater than 1.3%. The evperitnental error uct, 2.952 g., was dissolved in boiling petroleurn ether (60in determining pure cyclohexenecarboxylic acid by this 75') leaving behind only a trace of polymeric material. Filtration through a layer of Norit completely decolorized method is 2-3%. Neutralization equivalent alld alkaline the light yellow solution. On cooling and standing a t room hydrolysis to give bromide ion both corresponded to a prodtemperature, clusters of colorless elongated prisms formed. uct, C7H11Br02. After recrystallization from formic acid, This first crop after filtration, washing and drying weighed a sample was submitted for microanalysis. a sample mixed with trans-80.877 g., m.p. 115-116'; Anal. Calcd. for C7HIIBrOl: C, 40.59; 13, 5.35; Br, bromoacrylic acid melted 80-107". Subsequent crops 38.59. Foundz3: C, 40.85, 40.69; H , 5.47, 5.36; Br, hrought the total yield of reasonably pure product, m.p. 38.43, 38.40. 114-116', to 2.184 g. (79.5%). A second petroleum ether Isomerization Experiments. cis- and trans-2-Bromocyrecrystallization of the material was effected with a 96% clohexanecarboxylic Acids with 30% HBr-Acetic Acid.recovery in three crops, m.p. 115.0-115.6° cor. Further 2.20 g. (0.0126 mole) of each isomer was sealed in separafe recrystallizations did not improve this value. tubes with 13.6 g. (0.0504 mole) of 30% HBr in acetic acid Anal. Calcd. for C7H9Br02: C, 40.99; H , 4.42. Found%: and maintained a t a temperature of 100-110' for 24 hours. C, 41.28; H , 4.54. After cooling, the contents of the tubes (which had darktrans-2-Chlorocyclohexanecarboxylic Acid (VIIa).-H y- ened somewhat) were washed into a flask with acetic acid and then partially neutralized with 0.05 mole of sodium drogenation of 23.18 g. (0.1443 mole) of trans-2-chlorocyclohexene-4-carboxylic acid in acetic acid over 0.2 g. of Adams bicarbonate. The precipitated solid was dissolved with water and then the water and acetic acid were removed by catalyst gave, in four crops from petroleum ether (6!-75'), The resulting oily solid was dis0 Re- distillation a t ~ 1 2 mm. 22.26 g. (94.4%) of white crystals, m.p. 85-86 peated recrystallizations of the principal product raised the solved partially in 60 ml. of water and the remaining oil was taken up in 50 ml. of chloroform. The aqueous layer was melting point to 84.0-85.7' cor. Anal. Calcd. for C7H1lC1O2: C, 51.70; H, 6.82. Foundz5: washed with two 25-mI. portions of chloroform and the combined chloroform extracts were decolorized by passage C, 51.93; H, 6.83. through a 0.25-inch layer of Norit, dried, and concentrated. trans-2-BromocyclohexanecarboxylicAcid (VIIb).-HyIn the case of the cis isomer, 1.87 g. (85%) of material, drogenation of 23.716 g. (0.115 mole) of trans-2-bromocyclo- m.p. 74.5-98.0", was recovered (a crude phase diagram inhexene-4-carboxylic acid in 175 ml. of acetic acid over 0.2 dicates a t least 30% trans for this m.p.). The trans isomer g. of Adams catalyst afforded 23.62 g. of crude VIIb which yielded 1.82 g. ( 8 3 % ) , m.p. 90-101' (no more than 10% was dissolved in boiling petroleum ether (60-75') and was cis, from a crude phase diagram). After five recrystalhzacompletely decolorized by filtration through a layer of Norit. tions from 60-75' petroleum ether, the product from the Concentration and cooling of the filtrate afforded 20.32 g. cis-acid yielded 0.30 g. (16% of recovered material, 13.6% of white crystalline solid, m.p. 108-109". Subsequent based on original sample) of nearly pure trans-acid, m.1,. crops brought the total yield of reasonably pure VIIb to 107.0-108.5", giving no depression with pure trans-acid, 23.0 g. (96%). A second petroleum ether recrystallization and m.p. 86-103" with authentic cis acid. Similarly, three raised the melting point t o 109.0-109.8° cor. sample recrystallizations of the trans product gave 0.58 g. (31.8% mixed with the starting material melted a t 98-106 . of recovered, 26.4% of original) of nearly pur: trans-% Anal. Calcd. for C7HllBr02: C, 40.59; €I, 5.35. bromocyclohexanecarboxylic acid, m.p. 106-108 , 106.5Foundz6: C, 40.80; H , 5.46. 108.0' when mixed with pure trans isomer, 81-102" when cis-2-ChlorocyclohexanecarboxylicAcid (VIIIa).-A com- mixed with pure cis isomer. cis-2-BromocyclohexanecarboxylicAcid with Acetic Acid bustion tube containing 4.26 g. (0.0338 mole) of cyclohexand Potassium Bromide.-A mixture of 0.75 g. (0.00363 enecarbosylic a~idZ~p2~ and 12 g. of glacial acetic acid was mole) of 2-bromocyclohexanecarboxylic acid and 0.432 charged with anhydrous hydrogen chloride (8.0 g., 0.22 mole) and the tube was sealed and heated a t 100-110" for g. (0.00363 mole) of potassium bromide was placed in 23 15 hours. Upon opening, the reaction mixture was de- ml. of glacial acetic acid and allowed t o reflux 24 hours. colorized completely by filtration through a 0.25-inch layer After cooling, most of the potassium bromide (not all of of Norit. The solvent and excess hydrogen chloride were which had gone into solution) was filtered off and the acetic removed as completely as possible using an aspirator and acid removed in an air stream. The resulting solid was warming the reaction mixture at 70-80". The almost white triturated with 30 ml. of boiling petroleum ether and filqolid residue was stored over powdered potassium hydroxide tered. The solvent was removed in an air stream t o give 0.49 g. (65.4%) of a white, waxy solid, m.p. 91-99' (ind/in an evacuated desiccator for 12 hours; 5.33 g., m.p. 75cating no omore than 15% trans; mixed m.p. with pure czs 85', with only a slight odor of acetic acid. This material 104.5-109 ; mixed m.p. with pure trans 79-104') after was dissolved in about 40 ml. of boiling petroleum ether (60-75") and processed as for VIIb, several crops being standing in a vacuum desiccator 48 hours. cis-2-BromocyclohexanecarboxylicAcid with Acetic Acid. collected, m.p. 94-97', 3.62 g. One recrystallization of the combined crops raised the melting point t o 97.6-98.4' cor. The acid, 1.04 g. (0.0050 mole) of the acid (0.0050 mole) was refluxed 24 hours in 25 ml. of glacial acetic acid after whlch with almost quantitative recovery; 3.60 g. (66.3%). Anal. Calcd. for C7HIIClOz: C, 51.70; H, 6.82. the solvent was removed in an air stream to yield 0.70 g. (67.3%) of waxy solid, m.p. 100.5-110.2" (90% cis) after Foundta: C, 51.70; H, 6.87. three days in a vacuum desiccator. (25) Microanalysis by Coli Kodama, University of Michigan. ANNARBOR,MICH.
.
(26) F. Stolz, Ber., 19, 510 (1886).
(27) E. Boorman and R. P Linstead, J. Chcm. Soc., 258 (1935). (28) R. Van Coillte, Bull. soc. chim Belgcr, 45, 419 (1933).
(29) 0. Aschan. A n n . , 271, 274 (1892). (30) H.T.Bucherer and K. Dah1ern.J. p r o k t . C h r m . , 140,2(\5 ( 1 9 3 4 ) .
